The qt_feature calls are for "jpeg" and "png".
Pick-to: 6.5 6.6 6.7
Fixes: QTBUG-120036
Change-Id: I43e69d9fd1e08383b5ddfbfe194eaa2f2ae52f50
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Avoid some calls to detach() and use std::array instead raw c arrays.
Pick-to: 6.7
Change-Id: I95cbfb130b6fe2509e5d87184c6612718b90dd63
Reviewed-by: Axel Spoerl <axel.spoerl@qt.io>
CreateCursor() expects a ddb and according the docs a ddb must be
word-aligned but we created a byte-aligned ddb which results in a
wrongly rendered cursor later on.
Pick-to: 6.7 6.6
Task-number: QTBUG-82434
Change-Id: Ie847865a0d868043571faf4b7f4fdf6a92fce448
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
Also port from qMakePair() to just braced initialization and CTAD.
As a drive-by, use emplacement instead of appending pairs explicitly,
rewrite typedef into `using`, and remove the qpair.h include from the
.cpp file (cannot affect other TUs and <utility> is considered
included in qcompilerdetection.h).
Pick-to: 6.7
Task-number: QTBUG-115841
Change-Id: Ia985520dfce6b77b1c0fe7669fc92d2df97e1b06
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
During drag and drop operation, we used to get window under mouse and
process events for that window after the patch set
9ff5b886fade489b7cc268626f518145650f6b2c. But this sometimes cause hang
issue in GetMessage() when the respective window doesn't capture mouse
event and we are required to process mouse events.
This patch fixes this issue by considering whether the window under
mouse has capture.
Fixes: QTBUG-115260
Pick-to: 6.7 6.6 6.5 6.2
Change-Id: I748936e4ce23e60abce51d34c1326cfb105b06cb
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
Before this, when trying to build iOS tests, CMake could not find the
`LaunchScreen.storyboard` in the source directory, and therefore the
configuration was failing.
In addition, QtPublicFinalizerHelpers module was missing in QtBuild,
resulting in another configuration failure when configuring iOS tests.
Pick-to: 6.7
Change-Id: I592121892a2716973a92ec044414fa729fd3b15f
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Qt's UWP port has been dropped for Qt 6.0.
Pick-to: 6.7 6.6 6.5 6.2
Change-Id: I72f2cfbd1234110a0bd0c15669fa326328d4d08c
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
In the case when our foreign window is created to act as a parent for
embedding Qt content into a native Android app, we need a more "hands
off"
approach. If the foreign window represents an Android view which the
user
has created outside of Qt, we should not try to reparent it since it
already has a parent in the view hierarchy. Neither should we try to
remove it from the hierarchy, or set its visibility.
Pick-to: 6.7
Change-Id: Iea496578a40f45ebdd73947a1bb0e46a7131108c
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Add a helper method to tell whether the app is a standalone Qt for
Android app, where also the Context is created by Qt, or whether Qt
content is embedded as a View into a Context created by the user, i.e.
a "native" Android app.
Pick-to: 6.7
Change-Id: I618ba1c8cb40c9b132fc12a7ee6d54c071efa983
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
The SurfaceView class is not the best option for when we
have multiple windows, as the created Surface can only
either be below the window or on top of it, it is not a
part of the view hierarchy.
Replace the SurfaceView with TextureView when there are
more than one window. This way the surface will be a part
of the view hierarchy, and will respect z-ordering.
When there is only one window, keep using the SurfaceView
approach to limit the effect on existing apps, as well
as enable some of the benefits SurfaceView has for e.g.
game and multimedia apps, such as HDR ability.
Move touch handling from QtSurface to QtWindow, so
touches are handled also when using TextureView instead
of QtSurface aka SurfaceView.
Pick-to: 6.7
Task-number: QTBUG-118142
Change-Id: I37dcaf0fb656cfc1ff2eca0a3bfe7259f607411c
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
In the touch dispatcher methods, all events were routed for the
top level window at the position, which lead to all the events
being delivered for the parent window.
Since the JNI methods already have a parameter for window ID, take
that into use and determine the window to deliver the events to
by ID.
Pick-to: 6.7
Change-Id: I07ad7851c32a3e633ee748036d6818b6d0fa5588
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Previously, winId() returned a simple integer ID, incremented by one
for each window, instead of a handle to the underlying window.
Note, if constructing a QWindow with fromWinId(), the passed jobject
will not be the same one returned by winId(), as the passed jobject,
assumed to be a View, will be wrapped inside a QtWindow to ensure
child windows can be added to it if needed.
Pick-to: 6.7
Change-Id: I9d5d977eeb08d4cc3594f7339660fe94c3a55864
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Also port qMakePair() to just braced initialization and CTAD.
Pick-to: 6.7
Task-number: QTBUG-115841
Change-Id: I9eafb20ebc63bd33eb52661f4f14ad2db9fc0f00
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
The clear() will not preserve capacity, since the QList has been
copied into proxy_intervals already, so is shared. But then creating a
new QList is just as efficient, or more so.
Pick-to: 6.7
Change-Id: I7ffd91bae8ded5ac28d58bcc545c479f6edc7c90
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Also port from qMakePair to just braced initialization and CTAD.
Task-number: QTBUG-115841
Pick-to: 6.7
Change-Id: Ifd7220ae88b101351e91d1488ce18715eb4880e5
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
... with properly-named members.
This is in preparation of adding a new member in order to fix
QTBUG-101141 (namespaced base classes).
Pick-to: 6.7 6.6 6.5
Task-number: QTBUG-101141
Change-Id: I2309e425ac94ad275b54a898fd33a2891e5d1453
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
This code never worked:
For the comparison with max() - 32 to trigger, on 32-bit platforms (or
Qt 5) signed interger overflow would have had to happen in the
addition of the two sizes. The compiler can therefore remove the
overflow check as dead code.
On Qt 6 and 64-bit platforms, the signed integer addition would be
very unlikely to overflow, but the following truncation to uint32
would yield the correct result only in a narrow 32-value window just
below UINT_MAX, if even that.
Fix by using the proper tool, qAddOverflow.
Pick-to: 6.7 6.6 6.5 6.2 5.15
Change-Id: I7599f2e75ff7f488077b0c60b81022591005661c
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Putting the variable on the LHS of a relational operation makes the
expression easier to read. In this case, we find that the whole
expression is nonsensical as an overflow protection, because if
name.size() + value.size() overflows, the result will exactly _not_
be > max() - 32, because UB will have happened.
To be fixed in a follow-up commit.
As a drive-by, add parentheses around the RHS.
Pick-to: 6.7 6.6 6.5 6.2 5.15
Change-Id: I35ce598884c37c51b74756b3bd2734b9aad63c09
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
It was forgotten when we originally fixed the hardcoded 1_1 version.
Pick-to: 6.7 6.6 6.5
Change-Id: Iff3148d79466dac2830fe9a63d954aead96fb0ac
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
When a non-raster widget is inserted into the widget hierarchy, widgets
switches into using RHI for flushing its backingstore, matching the
window's surface type with the non-raster widget that was inserted.
This happens for example when inserting an QOpenGLWidget, which switches
the surface type to QWindow::OpenGLSurface, or a QQuickWidget, which
switches the surface type to that used for Qt Quick.
In these cases, there might still be child widgets with their own
window handle, and as long as these child windows have the same
surface type as the window owning the backing store these child
windows are flushed through the same rhiFlush() code path as the
top level window.
For child windows with a different surface type, QWidgetRepaintManager
chooses the pain QPlatformBackingStore::flush(). Unfortunately, the
implementation of QRhiBackingStore::flush() bails out when flushing
child windows, so when we were using QRhiBackingStore on macOS for
non-raster surfaces we would end up not flushing these child windows.
The way to fix this is to use QCALayerBackingStore for the case
where we know the client of the backing store (QWidgetWindow) is
going to call rhiFlush() explicitly for non-raster surfaces. That
will lead us into QCALayerBackingStore::flush() for the child windows
with a non-matching surface type, typically QWindow::RasterSurface,
and we'll flush those child windows by setting the layer's contents
property.
Fixes: QTBUG-119309
Pick-to: 6.7 6.6 6.5
Change-Id: Ia7052b9e651d575d8e34942a25e17ab5a1f5ce05
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
This patch replaces the QBasicFutureWatcher that was used for
continuations with context objects with a smaller QObject-based wrapper
that works directly from the actual continuation.
The idea stays the same: In order to run continuations in the thread of
a context object, we offload the continuation invocation to the
signal-slot mechanism.
Previously, we've hooked into QFuture with QFutureCallOutInterface to
emit a signal and trigger continuation invocation. However, it is much
easier and robust to emit that signal from the continuation itself.
This sidesteps the locking issues that QFutureCallOutInterface handling
presents. QFutureCallOutInterface basically requires any consumer to
only access the QFuture after all events have been posted and the
internal mutex unlocked, i.e. on the next cycle of the event loop.
Continuations do not impose this restriction; runContinuation()
explicitly unlocks the internal mutex before calling the continuation.
This fixes a deadlock when using QFuture::then(context, ...) where
the paren future is resolved from the same thread that the context
object lives in.
Fixes: QTBUG-119406
Fixes: QTBUG-119103
Fixes: QTBUG-117918
Fixes: QTBUG-119579
Fixes: QTBUG-119810
Pick-to: 6.7 6.6
Change-Id: I112b16024cde6b6ee0e4d8127392864b813df5bc
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Added a return type where it was missing as Clang would otherwise fail
to parse the command.
Qualified certain non-type template parameter that were based on a using
declaration inside the QSpan class.
An `\fn` command is parsed as a fake out-of-line definition. In that
context, the unqualified name will not be able to be solved by Clang,
resulting in an inability to match the element.
Pick-to: 6.7
Change-Id: I929ab5e8df4f7a6ef5d913266b584cd0393074e3
Reviewed-by: Topi Reiniö <topi.reinio@qt.io>
Since version 6.7, QDoc requires accurate definition of template
arguments in signatures passed to the \fn command.
Qualify the 'if_compatible_callback' template with the class it's
declared in.
Pick-to: 6.7
Change-Id: Ifd4618071c950e812c4df9a2e4e212317a6c9118
Reviewed-by: Luca Di Sera <luca.disera@qt.io>
The Qt::WindowState enum is documented to specify the current state of
top-level windows, and does not make sense for child windows. It's up to
the client code to ensure any sizing of child windows/widgets, e.g. via
layouts.
Many of our QPA backends bail out on QPlatformWindow::setWindowState()
for a child window. Ideally we would do this at a higher level, in
QWindow, but that's a bigger task, and the semantics of what happens
when a window is moved from being top level to child or back are not
fully clear.
As a first step, we ensure that the default window state when showing
the window via show() is Qt::WindowNoState.
Unfortunately, the QPlatformIntegration::defaultWindowState() API only
takes into account the window flags, which sadly do not have a way to
reflect whether the window is a child window or not (Qt::SubWindow is
not a child window, see QTBUG-115729).
We don't want to pass a QWindow to this API, as it would mean QWidget
would need to create a window when requesting the default window state.
Instead we hard-code the opt-out for child windows/widgets.
[ChangeLog][Gui/Widgets] Child windows and widgets are now always shown
in their normal state by show().
Pick-to: 6.7
Change-Id: Ie8caf2df2d854194b9985c9427a9eac1312aba65
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
This contains test specific variables like pid, sdkversion,
isPackageInstalled and isTestRunnerInterrupted which don't strictly
fall under 'Options'.
Pick-to: 6.7
Change-Id: I7c4a3422813f464b407d9b2be98c5b4a807c4f1e
Reviewed-by: Axel Spoerl <axel.spoerl@qt.io>
... from CMake so that Coin doesn't end up killing the whole VM if aRows
test gets stuck and instead allow androidtestrunner to cleanup and fetch
the logs so we know what happened, and even potentially ending up re-run
the test and succeeding if it was flaky.
Also, since CMake sets the timeout during configuration time, coin needs
to set COIN_COMMAND_OUTPUT_TIMEOUT under the build target where tests
are configured, so this moves the setting of that env var from the test
target to the build target.
Pick-to: 6.7 6.6 6.5
Change-Id: I9883ea1e98c93f79a088067518d09ca8acd5fdfd
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Leave a log message to know when the test runner timed out or the test
finished.
Pick-to: 6.7 6.6 6.5
Change-Id: If56ecaa5b0e3af22b3e26480a584e2d52ac97553
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
If the test runner was interrupted, stop waiting for the test
to start or to finish, go fetch the available test results
and logcat and uninstall the test app.
Also, set CMake TIMEOUT_SIGNAL_NAME to SIGINT and
TIMEOUT_SIGNAL_GRACE_PERIOD to 10 seconds to allow enough
time to fetch the logs and uninstall the test app.
Task-number: QTBUG-106479
Pick-to: 6.7 6.6 6.5
Change-Id: I4820cfe58f05d15179b4af819caa92e475881634
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
When QFileDialog::DontConfirmOverwrite is in the false state (default state),
it requires QFileDialog::AcceptMode to be set to QFileDialog::AcceptSave.
This information was not stated in the documents and this fix adds this
information to the docs.
Fixes: QTBUG-49720
Pick-to: 6.5 6.6 6.7
Change-Id: I1017252980810e712f2710751c8852dd12eaeef7
Reviewed-by: Topi Reiniö <topi.reinio@qt.io>
There is no need for {}, and removing them resolves the warning.
Change-Id: Id9812a9ba328f03839dd953662a82b05f99b0b8b
Reviewed-by: Christian Ehrlicher <ch.ehrlicher@gmx.de>
Properly delete the handle created by CreateDIBSection() with
DeleteObject() in case of error.
Pick-to: 6.7 6.6 6.5
Fixes: QTBUG-18057
Change-Id: Ibf40f6afb255bc5a7d7bf39939f5eb3dfcd19f84
Reviewed-by: Axel Spoerl <axel.spoerl@qt.io>
This amends 58907dfa81,
which tried to make sure we select a color variant of a
character when combining it with the VS-16 symbol, the
emoji selector (and vice versa with VS-15).
This is something which is not well-handled in Qt, and
the original work-around assumed that any font supporting the
conversion of Unicode characters to color glyphs would include
the VS-16 character in its map. This was true for the symptomatic
font on macOS, so it solved the problem there.
However, if you manually select a color font and then request a color
glyph, the font system should just pass it on and accept this. No
need for any magic adaptation. Same if you select a non-color font
and use VS-15.
To avoid regression, we detect this case and disable the behavior
introduced in 58907dfa81 when the existing
font selection is already meeting the criteria.
This is admittedly very hacky, and the plan is to do a larger refactor
where sequences of emojis are separated out ahead of time so that the
font selection can pick from a curated emoji font list for these.
But to fix the immediate regression, this will basically just return
to previous behavior for some cases which are currently prone to break,
so despite the added layer of conditions, it does improve the situation
temporarily.
Pick-to: 6.5 6.6 6.7
Task-number: QTBUG-111801
Fixes: QTBUG-118642
Change-Id: I1a3ca711426a042b22003b887883646132335a9f
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Creating an icon engine is a potentially expensive operation involving
many file lookups. In the case where neither QIconLoaderEngine nor the
engine from the platform theme is valid, QIconLoader tries to unsuccesfully
find the icon in all theme directories, the theme engine is constructed
potentially doing some expensive operation and finally a new
QIconLoaderEngine is constructed which does all the file lookups again.
Instead keep the existing QIconLoaderEngine around if it was already
constructed.
Pick-to: 6.7
Change-Id: Iace9a3f904730064f44939b2269316484ac6da2e
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Amends 32edae5e26, after which we keep a
copy of the restored state if the state couldn't be applied yet. Since
making a copy of the entire state results in multiple copies of layout
item pointers, we might end up with dangling pointers if the layout
structure is modified while we keep the copy. This can happen if methods
such as tabifyDockWidgets or splitDockWidget get called; e.g. tabifying
dock widgets will destroy the layout items that were added for them.
Unfortunately, the layout items do not have a pointer back to the layout
they live in, and the items in the stored state might not yet live in a
layout anyway. So we cannot remove the items from their layout in a
QDockWidgetItem destructor implementation.
Instead, we have to forget the stored state. Add a helper function that
writes the stored state back to the actual state, and deletes the stored
state afterwards. Call this function when the layout might get modified
programmatically.
Add a test case that reproduces the crash without the fix, and passes
with the patch.
Fixes: QTBUG-120025
Pick-to: 6.7 6.6 6.5
Change-Id: I8f7e886f3c4ac38e25f9b8bc194eea0833e5974f
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Axel Spoerl <axel.spoerl@qt.io>
... instead of just choosing the closest scaling factor.
Otherwise we end up with the 1x pixmap for a 1.25 scaling factor instead
of a the better fitting 2x pixmap.
Task-number: QTBUG-90634
Change-Id: Ic554fc8d2715deea43bc22e71414902a263b2fef
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Now that the assignment-bitwise operators can reuse storage, we can make
these operators also be capable of reusing storage.
Pick-to: 6.7
Change-Id: I85b3fc2dd45c4693be13fffd1795b893de65a5b8
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
Instead of creating a temporary copy of one of the two sides (which will
share QByteArray), create one with the correct target size such that it
is already detached.
Drive-by move them to hidden friends.
Pick-to: 6.7
Change-Id: I85b3fc2dd45c4693be13fffd1795b74eeaf3be71
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
1b96c75661f678632485513d1323166bc6d74e5c reduced event noise by
dropping accessibility events if Windows UI automation has not been
activated; however, this did not take into account that some widgets
(e.g. QMessageBox) rely on accessibility event processing to emit
certain system sounds.
Change notifyAccessibilityUpdate() so that system sound related events
are processed regardless of accessibility activation.
Pick-to: 6.5 6.6 6.7
Change-Id: I22f1516e8fbb3727b065ecc2c30b272b2d6fd1c0
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
QPair _is_ std::pair. It's confusing that debuggers consistently show
the real name, std::pair, while the API and docs continue to maintain
the illusion that there is such a thing as QPair.
Use std::pair everywhere.
Pick-to: 6.7 6.6 6.5
Task-number: QTBUG-115841
Change-Id: I009e2fc415a79a74b583a13cf11e4ff9483a7f6b
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
Also port qMakePair() to just braced initialization and CTAD.
Pick-to: 6.7
Task-number: QTBUG-115841
Change-Id: I46ee214ab47513375a6e28e3b439c7b060581235
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
Remove the __FILE__ preprocessor macro that contains the full path to
the current source file. This is needed for reproducible builds.
The macro was only used for two debug messages. It's enough to mention
the file name - if at all.
Fixes: QTBUG-96281
Pick-to: 6.5 6.6 6.7
Change-Id: I7940ec090b570ec0105196a881e4c73c279bd248
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Also port from qMakePair to just braced initialization using CTAD.
Task-number: QTBUG-115841
Pick-to: 6.7
Change-Id: Ib0ad55d7110521e34004dc9050022f9c0046722e
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
The original is in qcontainerfwd.h these days, so just include that.
Pick-to: 6.7
Change-Id: I1d0adaf528f158c48caf47d65b20d09eff06f57d
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Amir Masoud Abdol
Christian Ehrlicher
Marc Mutz
Santhosh Kumar
Oliver Wolff
Tinja Paavoseppä
Mårten Nordheim
Tor Arne Vestbø
Arno Rehn
Luca Di Sera
Topi Reinio
Assam Boudjelthia
Safiyyah Moosa
Audun Sutterud
Eskil Abrahamsen Blomfeldt
David Redondo
Volker Hilsheimer
David Schulz
Thiago Macieira
Christian Heimlich
Joerg Bornemann